INDEX

Explanations

specific names and references associated with academic research

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 11-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L11R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.11.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

kla

-0.17

 smoke

-0.15

arro

-0.15

Ð¾ÑĢÐ°Ð»ÑĮ

-0.15

Å¥

-0.15

è¤

-0.14

 smoking

-0.14

seg

-0.14

sass

-0.14

otte

-0.14

POSITIVE LOGITS

 dane

0.14

ħ§

0.14

oloj

0.13

-Token

0.13

-valu

0.13

 wool

0.13

 jsonResponse

0.13

 cock

0.13

olean

0.13

undy

0.12

Activations Density 0.003%

specific names and references associated with academic research

No Comments

No Known Activations